[Tlhingan-hol] Interactions between verb suffixes

[SuStel]

On 12/21/2015 2:07 PM, Will Martin wrote:
>> I wonder if we can re-categorize these suffixes according to their TKD
>> explanations for examination (ignoring rovers).
>>
>> Describe subject:
>> -nIS, -qang, -rup, -beH, -vIp, -moH, -lu' -laH
>> (Types 2, 4, 5)
>
> Describe the subject:
> -nIS, -qang, -rup, -beH, -vIp, and -laH
> (Type 2, and half of the Type 5 suffixes)
>
> Reassigns the agent of action of the root verb, my term for what is
> called the subject anywhere in TKD, to the syntactic role of object, so
> that a different noun becomes the syntactic subject in the resulting
> construction:
>
> -moH, -lu’
> (Type 4 and the other half of the Type 5 suffixes)

But those two suffixes STILL describe the subject of the sentence. 
{-moH} says the subject causes, but does not perform, the action; {-lu'} 
says the subject is indefinite.

What you call the "agent of action of the root verb" is simply known as 
an "agent." That is, the entity that performs the action that's 
happening. In {tlhIngan HeghmoH Human}, the Klingon is the agent, not 
the human. The action is "die," not "cause." (Semantics is actually more 
detailed than this, but for purposes of this discussion we can limit the 
term to just "agent.")

{-lu'} does not reassign roles to any nouns:

    tlhIngan Hol ghoj Human
    [ patient  ]      [agent]

    tlhIngan Hol ghojlu'
    [ patient  ]

The verb prefixes are used differently than in sentences with a definite 
subject, but the object remains the same. I see no reason to exclude 
{-lu'} from the category of suffixes that describe the subject.

{-moH} may be "weird" in how it reassigns semantic roles in the 
sentence; or the volition/predisposition suffixes, and maybe {-laH}, may 
instead simply be tied to semantic roles instead of the syntactic one 
described in TKD (or both). If this is the case then {-moH} does not 
deserve a special category of its own; all it's doing is saying "the 
subject causes a change; shift some semantic roles to support this."

> Yes, this is messy. It breaks down the suffixes according to boundaries
> that don’t match the boundaries of suffix Type. This is part of what
> makes it messy. The whole reason that the pair {-lu’} and {-laH} is the
> most common type that tempts us to use two suffixes of the same type,
> enough that there’s a slang suffix {-luH} to handle that conflict, and
> it’s the ONLY such slang suffix that Okrand has acknowledged.

Well, or its counterpart {-la'}.

> In particular, this is where {-moH} introduces the idea you like to call
> “ditransitive” even though Okrand completely avoids the term
> “transitive” at all.

I tend to avoid that word. Are you referring to some other "you" than me?

> Given canon, it seems like that any verb that has a typical capacity of
> involving a subject and an object, once {-moH} is added, moves the
> explicit subject to the object slot if there is no explicit object, and
> gets further moved and appended with {-vaD} if there is an explicit
> object, even though he never indicates that nouns with {-vaD} are any
> kind of object. They are “beneficiaries”, which is qualitatively
> different from “objects”.

Hence my age-old theory of a "header" position. The header is the space 
where all the nouns that are not subjects or (direct) objects go. You 
can have as many as you need. Sometimes they're marked with syntax 
suffixes; sometimes they're unmarked time stamps. There is no order 
specified in the header. And there are cases where nouns marked with 
syntax suffixes become subjects or objects and are NOT headers.

> And the ONLY time that {-vaD} is optional for a noun is when it is the
> agent of action of a verb that has an agent of cause also has an
> explicit direct object.

When there's a causer, a third-person agent, and patient, the causer is 
the subject, the patient is the object, and the agent is a header marked 
as a beneficiary and considered an indirect object.

When there's a causer, a first- or second-person agent, and a patient, 
the speaker has the option to use the above OR to elide the pronoun and 
use the verb prefix to specify the combination of causer and agent. This 
is what I called the prefix trick.

> All of the following describe the same situation:
>
> QeD Daghoj. You learn science.
>
> qaghojmoH. I cause you to learn.
>
> Here’s where it gets strange:
>
> QeD vIghojmoH.
>
> “I cause science to learn”?!
>
> QeD qaghojmoH.
>
> “I cause you to learn" — and “science” is oddly placed in the position
> of object of causing to learn, just as we would place things in {Sor
> vImeQmoH} for “I cause the tree to burn”.
>
> SoHvaD QeD vIghojmoH.
>
> “I cause science to learn for your benefit.” ?!

They all describe different aspects of what could be the same situation. 
These sentences are not interchangeable.

QeD Daghoj
says nothing about how or why you learn science.

qaghojmoH
says nothing about what I'm teaching you.

QeD vIghojmoH
says nothing about whom I'm teaching science to.

QeD qaghojmoH
SoHvaD QeD vIghojmoH
say the same thing, and include all roles of causer, agent, and patient.

> And nobody understands why I think this is ugly.

*shrug* I think, with regard to languages, what you call "ugly" I call 
"the way it works" and "natural." Language is a slippery thing; it can't 
be set down once and for all in a rule book.

>> Describe action:
>> -choH, -qa', -pu', -ta', -taH, -lI'
>> (Types 3, 7, {-Ha'})
>>
>> Express speaker's position:
>> -chu', -bej, -law', -ba', -neS
>> (Types 6, 8)
>>
>> Syntax:
>> -DI', -chugh, -pa', -vIS, -bogh, -meH, -'a', -wI', -mo'
>> (Type 9)
>>
>> Interaction of subject, object, and action:
>> -'egh, -chuq
>> (Type 1)
>
> I feel like this descriptor misses the primary point of Type 1: That the
> subject and object become the same entity, either individually (for
> members of a group) or wholly. That’s the “each other” vs. “self”
> meaning that requires two different Type 1 suffixes.

I know what the suffixes MEAN, but we're interested in how suffixes 
interact with each other when defined in terms of what role they affect 
grammatically. The definition of {'egh} says "the action described by 
the verb affects the performer of the action, the subject." It's tied 
into both action and subject. We're also told that such verbs take 
"no-object" prefixes. All this tells me that {'egh} (and I presume 
{-chuq}, which has no grammatical explanation in TKD) has a more 
complicated impact on the structure of the sentence than the other 
suffixes do, and this is indeed what we find. Sure, {-moH} may have been 
confusing before we got lots of examples, but the Type 1 suffixes are 
downright restrictive, and seem to change how they work depending on the 
meaning of the verb to which they're attached, not the suffixes they're 
found alongside.

Either way, they really belong by themselves.

-- 
SuStel
http://trimboli.name